It uses a couple germanium pnp's
I "cleaned" up the schematic and breadboarded this a while ago
It was cool, but I didn't spend enough time to get it to the point where it was worth boxing.
So, I breaded it again. I used some "standard" FZ-1A compatible transistors since it does resemble the 2nd and 3rd stages of that....used Q1 ~ 75 hfe, leak ~ 150uA and Q2 ~ 120 hfe, leak ~ 320uA
I got a good fuzz right off, though nothing like an FZ-1A. It's fatter and heavier. Q1C/Q2C voltages ~ -1.1V and -1.2V respectively.
A couple notes: definitely screw with different leakages in Q2. Just like many fuzzes with this structure, it's biased by leakage....and screw with the 1M on Q1 and 22K /10K on Q2 to tune it where you like it best.
The stock values gave me a good fuzz right off, but with slight gating and a swell. Increasing the 1M to 1.5M got rid of that, but also increases the bass response a bit. The 100K was too small of a "cutoff"...it's too dead there. I needed something in the 200K-300K range.
The 22K /10K on Q2 were pretty damn close, also. For example, I tuned the 22K by ear and got 28K for the best.
That was just my first shot. I'll definitely be experimenting with many other transistors and leakages, especially for Q2.